Atrial constitutive neural networks

2025-04-03

AI Summary

Traditional phenomenological constitutive models struggle to accurately capture the complex mechanical behavior of atrial tissue. To address this, we propose a data-driven constitutive neural network (CNN) method—applied here for the first time to biaxial tensile experimental data from both healthy and atrial fibrillation (AF) atrial tissues. Our approach integrates physics-informed constraints with tensor invariant embedding to rigorously enforce objectivity, enabling end-to-end, automatic discovery of physically consistent and interpretable constitutive relationships directly from experimental data. The resulting model achieves significantly higher fidelity in data fitting and superior generalizability across physiological and pathological states compared to classical hyperelastic models. This framework establishes a new paradigm for high-fidelity cardiac biomechanical simulation and arrhythmia risk prediction.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Machine Learning: Matrix & Tensor Methods

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

This work presents a novel approach for characterizing the mechanical behavior of atrial tissue using constitutive neural networks. Based on experimental biaxial tensile test data of healthy human atria, we automatically discover the most appropriate constitutive material model, thereby overcoming the limitations of traditional, pre-defined models. This approach offers a new perspective on modeling atrial mechanics and is a significant step towards improved simulation and prediction of cardiac health.

Problem

• Characterize atrial tissue mechanics using neural networks
• Discover optimal material model from biaxial test data
• Improve cardiac health simulation and prediction accuracy

Innovation

• Uses constitutive neural networks for atrial tissue
• Automatically discovers material models from data
• Overcomes limitations of traditional predefined models